GEOPHYSICAL RESEARCH LETTERS
Supporting Information for “Recent and future
changes in the Asian monsoon – ENSO relationship:
Natural or forced?”
Xiaoqiong Li,
1,2
1
and Mingfang Ting
Corresponding author: X. Li, Lamont-Doherty Earth Observatory, Columbia University, 61
Route 9W, Palisades, NY 10964, USA. ([email protected])
1
Lamont-Doherty Earth Observatory,
Columbia University, Palisades, New York,
USA.
2
Department of Earth and Environmental
Sciences, Columbia University, New York,
New York, USA.
X-2
LI AND TING: CHANGES IN ENSO-MONSOON RELATIONSHIP
Contents of this file
1. Text S1
2. Figures S1 to S3
3. Table S1
Introduction
This supporting material contains additional information on the signal-to-noise (S/N)
maximizing empirical orthogonal function (EOF) analysis used to extract the externally
forced signal of global sea surface temperature (SST) and separate the natural and forced
components of ENSO-monsoon relationship. Additionally, the list of Coupled Model
Intercomparison Project - Phase 5 (CMIP5) models and the number of realizations for
both the historical and representative concentration pathway 8.5 (RCP8.5) simulations
used in the study are provided.
LI AND TING: CHANGES IN ENSO-MONSOON RELATIONSHIP
X-3
Text S1.
Figure S1 shows the first EOF mode of S/N maximizing EOF analysis on June-August
(JJA) seasonal average global SST for historical 1901-2005 (a, c) and rcp8.5 2006-2099
(b, d). The percentage of variance explained is 56% for historical and 69% for rcp8.5.
The spatial structures (a, b), shown as regression patterns of SST anomalies onto the
leading principle components (S/N PC1), both display global warming patterns with high
model-agreement. The S/N PC1 for the 20th century (c) shows a gradual increase until
a drop in the 1960s, followed by a faster increase rate in the past few decades. This
temporal evolution of SST warming trend is consistent with Ting et al. [2009, 2011],
using CMIP3 models. However our result here shows more prominent fluctuations due to
the fact that we directly applied S/N maximizing EOF analysis onto JJA SST without
any prior smoothing. Some of these fluctuations are consistent with studies addressing
the effect of volcanic aerosols [Ammann et al., 2003; Meehl et al., 2004]. In the rcp8.5
case (d), the S/N PC1 time series exhibits an almost linear increasing trend throughout
the 21st century.
Figure S2 shows the forced Niño 3.4 SST index for each model (using the first realization)
and the observation using Extended Reconstructed Sea Surface Temperature (ERSST).
Figure S3 shows the total and natural Niño 3.4 SST indices for the observation (ERSST).
References
Ammann, C. M., G. A. Meehl, W. M. Washington, and C. S. Zender (2003), A monthly
and latitudinally varying volcanic forcing dataset in simulations of 20th century climate,
Geophys. Res. Lett., 30 (12), 1657, doi:10.1029/2003GL016875.
X-4
LI AND TING: CHANGES IN ENSO-MONSOON RELATIONSHIP
Meehl, G. A., W. M. Washington, C. M. Ammann, J. M. Arblaster, T. M. L.
Wigley, and C. Tebaldi (2004), Combinations of natural and anthropogenic forcings in twentieth-century climate, J. Climate, 17 (19), 3721–3727, doi:10.1175/15200442(2004)017<3721:CONAAF>2.0.CO;2.
Ting, M., Y. Kushnir, R. Seager, and C. Li (2009), Forced and internal twentiethcentury SST trends in the North Atlantic, J. Climate, 22 (6), 1469–1481, doi:
10.1175/2008JCLI2561.1.
Ting, M., Y. Kushnir, R. Seager, and C. Li (2011), Robust features of Atlantic multidecadal variability and its climate impacts, Geophys. Res. Lett., 38 (17), L17,705, doi:
10.1029/2011GL048712.
X-5
LI AND TING: CHANGES IN ENSO-MONSOON RELATIONSHIP
(a)
(b)
SST S/N EOF1 CMIP5 historical 26/34
60oN
60oN
30oN
30oN
0o
0o
30oS
30oS
60oS
60oS
60oE
120oE
180oW
−0.32 −0.24 −0.16 −0.08
0
120oW
0.08
60oW
0.16
0.24
SST S/N EOF1 CMIP5 rcp85 26/34
60oE
0.32
−2
−1.5
120oE
−1
180oW
−0.5
(d)
SST S/N EOF PC1 historical
3
2.5
2
1
1.5
2
SST S/N EOF PC1 rcp85
2
Variance = 0.69208
Variance = 0.56371
1.5
Noise EOFs = 57 (80%)
Standardized S/N PC1
1.5
Standardized S/N PC1
0.5
60oW
EOF spatial structure
EOF spatial structure
(c)
0
120oW
1
0.5
0
−0.5
−1
−1.5
Noise EOFs = 57 (80%)
1
0.5
0
−0.5
−1
−2
−2.5
1900
1920
Figure S1.
1940
1960
Years
1980
2000
−1.5
2010
2020
2030
2040
2050 2060
Years
2070
2080
2090
2100
First EOF mode of S/N maximizing EOF analysis of JJA global SST for CMIP5
(a,c) historical 1901-2005 and (b, d) RCP8.5 2006-2099 simulations. (a, b) Spatial structures,
shown as regressions of SST anomalies onto standardized S/N PC1, stippling denotes 26/34
model-agreement; and (c, d) standardized leading principle components (S/N PC1).
X-6
LI AND TING: CHANGES IN ENSO-MONSOON RELATIONSHIP
Nino 3.4 SST forced historical
(b)
MIROC−ESM
CanESM2
bcc−csm1−1−m
FIO−ESM
CESM1−BGC
BNU−ESM
CCSM4
MPI−ESM−LR
IPSL−CM5A−LR
IPSL−CM5A−MR
GFDL−ESM2G
GISS−E2−H
bcc−csm1−1
CMCC−CESM
CMCC−CMS
MPI−ESM−MR
GFDL−ESM2M
NorESM1−ME
NorESM1−M
GISS−E2−R
IPSL−CM5B−LR
MIROC5
MIROC−ESM−CHEM
CNRM−CM5
CESM1−CAM5
MRI−CGCM3
HadGEM2−CC
CMCC−CM
inmcm4
GFDL−CM3
ACCESS1−0
ACCESS1−3
CSIRO−Mk3−6−0
HadGEM2−ES
0.6
SST index [°C]
0.3
0
−0.3
−0.6
−0.9
1900
1920
1940
1960
Years
1980
2000
Nino 3.4 SST forced rcp85
3
GFDL−CM3
MIROC−ESM−CHEM
CSIRO−Mk3−6−0
MIROC−ESM
IPSL−CM5A−MR
CMCC−CESM
HadGEM2−ES
IPSL−CM5A−LR
ACCESS1−3
HadGEM2−CC
CanESM2
CMCC−CMS
CMCC−CM
ACCESS1−0
CESM1−CAM5
BNU−ESM
MPI−ESM−MR
bcc−csm1−1−m
MRI−CGCM3
bcc−csm1−1
IPSL−CM5B−LR
MPI−ESM−LR
GFDL−ESM2G
GFDL−ESM2M
GISS−E2−H
CNRM−CM5
FIO−ESM
NorESM1−ME
NorESM1−M
CCSM4
MIROC5
CESM1−BGC
GISS−E2−R
inmcm4
2
1
SST index [°C]
(a)
0.9
0
−1
−2
−3
2010
2030
2050
Years
2070
2090
Figure S2. Forced component of JJA Niño 3.4 SST index for CMIP5 (a) historical 1901-2005
and (b) RCP8.5 2006-2099 simulations, using the first realization for each model. The models are
ranked based on each model’s warming amplitude. Black thick line in (a) shows the observation
(ERSST). Units are ◦ C.
X-7
LI AND TING: CHANGES IN ENSO-MONSOON RELATIONSHIP
Nino 3.4 SST index ERSST
2
1.5
SST index [°C]
1
0.5
0
−0.5
−1
−1.5
−2
1900
Figure S3.
Total
Natural
1920
1940
1960
Years
1980
2000
Total and natural components of JJA Niño 3.4 SST index for the observation
(ERSST). Units are ◦ C.
X-8
Table S1.
LI AND TING: CHANGES IN ENSO-MONSOON RELATIONSHIP
List of CMIP5 models and the number of realizations used in the study.
Model
Institute ID
historical rcp8.5
ACCESS1-0
CSIRO-BOM
1
1
ACCESS1-3
CSIRO-BOM
1
1
bcc-csm1-1
BCC
3
1
bcc-csm1-1-m
BCC
3
1
BNU-ESM
GCESS
1
1
CanESM2
CCCMA
5
5
CCSM4
NCAR
6
6
CESM1-BGC
NSF-DOE-NCAR
1
1
CESM1-CAM5
NSF-DOE-NCAR
3
3
CMCC-CESM
CMCC
1
1
CMCC-CM
CMCC
1
1
CMCC-CMS
CMCC
1
1
CNRM-CM5
CNRM-CERFACS
10
5
CSIRO-Mk3-6-0
CSIRO-QCCCE
10
10
FIO-ESM
FIO
3
3
GFDL-CM3
NOAA GFDL
5
1
GFDL-ESM2G
NOAA GFDL
1
1
GFDL-ESM2M
NOAA GFDL
1
1
GISS-E2-H
NASA GISS
5
1
GISS-E2-R
NASA GISS
6
2
HadGEM2-CC
MOHC
1
1
HadGEM2-ES
MOHC
4
4
inmcm4
INM
1
1
IPSL-CM5A-LR
IPSL
5
4
IPSL-CM5A-MR
IPSL
1
1
IPSL-CM5B-LR
IPSL
1
1
MIROC-ESM
MIROC
3
1
MIROC-ESM-CHEM
MIROC
1
1
MIROC5
MIROC
5
3
MPI-ESM-LR
MPI-M
3
3
MPI-ESM-MR
MPI-M
3
1
MRI-CGCM3
MRI
3
1
NorESM1-M
NCC
3
1
NorESM1-ME
NCC
1
1